Relation Preserving Triplet Mining for Stabilising the Triplet Loss in Re-identification Systems

Object appearances change dramatically with pose variations. This creates a challenge for embedding schemes that seek to map instances with the same object ID to locations that are as close as possible. This issue becomes significantly heightened in complex computer vision tasks such as re-identification(reID). In this paper, we suggest that these dramatic appearance changes are indications that an object ID is composed of multiple natural groups, and it is counterproductive to forcefully map instances from different groups to a common location. This leads us to introduce Relation Preserving Triplet Mining (RPTM), a feature-matching guided triplet mining scheme, that ensures that triplets will respect the natural subgroupings within an object ID. We use this triplet mining mechanism to establish a pose-aware, well-conditioned triplet loss by implicitly enforcing view consistency. This allows a single network to be trained with fixed parameters across datasets while providing state-of-the-art results. Code is available at

PDF Abstract Proceedings of 2023 PDF Proceedings of 2023 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Person Re-Identification DukeMTMC-reID RPTM Rank-1 93.5 # 7
Rank-5 96.1 # 3
mAP 89.2 # 12
Vehicle Re-Identification VehicleID Large RPTM mAP 80.5 # 2
Rank-1 92.9 # 4
Rank-5 96.3 # 4
Vehicle Re-Identification VehicleID Medium RPTM mAP 81.2 # 2
Rank-1 93.3 # 4
Rank-5 96.5 # 4
Vehicle Re-Identification VehicleID Small RPTM mAP 84.8 # 2
Rank-1 95.5 # 3
Rank-5 97.4 # 6
Vehicle Re-Identification VeRi-776 RPTM mAP 88 # 1
Rank-1 97.3 # 1
Rank1 97.3 # 1
Rank5 98.4 # 1


No methods listed for this paper. Add relevant methods here